Welcome to GeoStep Documentation

GeoStep is a proprietary Python library for designing, simulating, and analyzing marketing experiments using Geographic Randomized Controlled Trials (Geo-RCTs).

This documentation provides a guide to using the library, from installation and getting started to understanding the underlying methodologies and exploring the full API reference.

Documentation Structure

Getting Started

• Installation: How to install the library and its dependencies

• Getting Started: Step-by-step tutorial for your first geo-experiment

• Business Integration Guide: ROI analysis, use cases, and integration strategies

Core Documentation

• Methodology: Statistical theory behind RCTs, Stepped-Wedge, and Staircase designs

• API Reference: Complete function and class documentation

• Advanced Topics: Mathematical foundations and sophisticated techniques

Support & Troubleshooting

• Troubleshooting Guide: Common issues, solutions, and best practices

Why GeoStep?

In a world where marketing accountability is paramount, GeoStep provides a scientifically rigourous framework to move beyond correlation and measure the true, causal impact of your marketing investments.

Key Benefits:

• Scientific rigour: Gold standard randomized controlled trials

• Causal Proof: Eliminate correlation vs. causation confusion

• ROI Optimization: 10-30% improvement in budget allocation efficiency

• Incrementality Focus: Measure true lift, not just total attribution

• Enterprise Ready: Production-grade tools for large-scale experiments

Integration with other tools

• GeoLift: Retrospective causal analysis

• MMM: Marketing mix modeling and attribution

---